l l - American Chemical Society

Madison, Wisconsin 53706. Receioed Nocember 5, 1966 grant from the National Science Foundation. Nonplanar Cyclobutane. I. The. 3-Tsopropylcyclobuty1 ...
0 downloads 0 Views 257KB Size
155

Acknowledgment. This research was supported by a grant from the National Science Foundation. David C. Zecher, Robert West Department of Chemistry, University of Wisconsin Madison, Wisconsin 53706 Receioed Nocember 5, 1966 1

Nonplanar Cyclobutane. I. The 3-Tsopropylcyclobuty1 System. Nuclear Magnetic Resonance Spectra of Alcohols and Amines’ Sir: During the past 15 years, a good deal of physical information has accumulated which supports a nonplanar structure for cyclobutane. 2-6 Recently, chemical evidence for a folded conformation for the cyclobutane system has been adduced from equilibration data.’ Cyclobutane is quite flexible and exhibits a dynamic ring-bending equilibrium which has been termed “pseudo-rotation.8 In a pair of conformers representing the extremes of such flexion, equatorial and axial configurations may be assigned to substituent^.^ We have attempted to utilize a bulky group in a 1,3disubstituted cyclobutane to retard this equilibrium to delineate these positions better in a manner analogous to the effect of “holding” groups in the cyclohexane system. Thus, in Figure 1, favored conformations

ris

I

I

1

1

1

I

L

I

I

I . . r

. . I

I

i . i Figure 2.

I

I

I

1 1

I ’

I/,// ll

&y

s R

4

1

I

I

Figure 3.

Figure 1.

appear on the left. We hoped in this fashion to probe possible differences in behavior between axial and equatorial substituents, perhaps paralleling differences in cyclohexanes, and to provide better definition of the condition of nonplanarity. Accordingly, we have prepared and examined a series of 3-isopropylcyclobutane derivatives. In this communication, we wish to present the nmr spectra of the isomeric 3-isopropyl(1) Presented at the 152nd National Meeting of the American Chemical Society, New York, N. Y . ,Sept 1966, Abstract S167. (2) A. Almenningen, 0. Bastiansen, and P. N. Skancke, Acta Chem. Scand., 15, 711 (1961). (3) J. D. Dunitz and V . Schomaker, J. Chem. Phys., 20,1703 (1952). (4) G . W. Rathjens, N. K. Freeman, W. D. Gwinn, a n d K . S. Pitzer, J . Am. Chem. SOC.,75, 5634 (1953). ( 5 ) R. C. LordandI. Nakagawa,J. Chem. Phys., 41,2951 (1963). (6) W. G . Rothschild and B. P. Dailey, ibid., 36, 2931 (1962); L. C. Snyder and S. Meiboom, Chem. Eng. News, 44, No. 17, 51 (1966). (7) N. L. Allinger and L. A. Tushaus, J . Org. Chem., 30,1945 (1965). (8) J. E. Kilpatrick, K. S. Pitzer, and R. Spitzer, J . Am. Chem. SOC., 69,2483 (1947); see also J. B. Hendrickson, ibid., 83, 4537 (1961). (9) For cases in which there is a single halogen substituent, recent work has indicated that the “axial” conformer may be relatively planar; see J. B. Lambert and J. D. Roberts, J . Am. Chem. SOC.,87, 3884, 3891 (1965); H. Kim and W. D. Gwinn,J. Chem. Phys., 44,865 (1966); W. G . Rothschild, ibid., 45, 1214 (1966). (10) S. Winstein and N. J. Holness, J . Am. Chem. SOC., 77, 5592 (1955).

cyclobutanols and 3-isopropylcyclobutylamines (Figures 2 and 3), which provide convincing and dramatic evidence for the model depicted. The spectra were recorded for the amines at 100 Mc and the alcohols at 60 Mc. In both sets of spectra, the unsymmetrical doublet of the methyl protons appears similarly centered at ca. 0.8 ppm (relative to TMS). In the amines, the amino protons may be seen as a sharp resonance at about 1.6 ppm for the cis and 1.8 ppm for the trans isomer. The proton geminal to the amino group is located as a multiplet centered at 3.2 ppm for the cis and 3.4 ppm for the trans isomer. The latter assignment is reasonable since these protons are the most strongly deshielded by the amino group and should appear at lowest field. This conclusion was verified by a double resonance experiment which showed this resonance coupled to the resonances centered at 2.3 and 1.3 ppm in the cis and at 1.9 ppm in the trans isomer. The latter resonances are due to the ring methylene protons, and it is here that a pronounced distinction between the cis and trans isomers is seen. (11) Full details of the synthesis and configurational assignments of the alcohols and amines will be published in a forthcoming paper concerned with deamination of the amines.

Communications to the Editor

156

A quite similar situation exists for the alcohols, The OH proton appears at 3.1 ppm for the cis and at 5.2 ppm for the trans isomer; the proton geminal to the OH is centered at 4.1 ppm for the cis and 4.3 ppm for the trans isomer. Ring methylene protons are centered at 2.4 and 1.4 ppm for the cis and at 2.1 ppm for the trans isomer. In the alcohols, as determined by relative integrated peak areas, remaining tertiary protons of the isopropyl group and C3 appear overlapped by the high-field methylene resonances. The same situation was observed in the amines. If our conformational picture of cyclobutane is a reasonable one, analogies with nmr effects in cyclohexane should be valid. The large variance between the cis and trans ring methylene signals is strongly reminiscent of the “axial effect” observed in the cyclohexane~.’~-’’ It has been observed in several instances that axial substituents cause a narrowing and intensification of the resonance due to equatorial and axial ring protons. Such cases as OH in the decalins’6 and steroidsI8 and alkyl groups in cyclohexane^^^ are illustrative. Equatorial protons ordinarily absorb at lower field than axial protons in cyclohexane. This distinction is actually seen only in temperature-frozen or conformationally homogeneous (i.e., t-butylcyclohexaneZ0) systems, since ring equilibration causes averaging. If, in a homogeneous disubstituted system, the shielding effects of substituents on axial us. equatorial protons were disparate but unopposed, one would expect broadening of overlapping resonances or even separation, whereas opposition would result in narrowing of the signal. Equatorial alkyl groups have been shown to shield cis (axial) adjacent protons more strongly than trans (equatorial), l9 while axial substituents (including OH and alkyl) shield cis (equatorial) adjacent protons and deshield trans (axial) adjacent protons. l 8 , l 9 Thus, in the trans 3-isopropylcyclobutane isomers, with predominantly equatorial isopropyl and axial OH or NH2 expected, effects are opposed, resulting in resonance narrowing and overlapping for the ring methylene signals. In line with previous reasoning, in the cis isomers equatorial OH or NH, groups, which are strongly electronegative, must deshield trans (equatorial) more than cis (axial) ring protons, so that the equatorial protons more strongly deshielded by equatorial OH or NH, appear further downfield than the axial protons more strongly shielded by the equatorial isopropyl group. A similar effect may be noted in trans-4-t-butyland cis-3-t-butylcyclohexylamine, which possess equatorial NH, groups and two separate ring methylene signals, whereas their isomers have single more intense bands at intermediate field.’j We have likewise ob(12) Relative areas going upfield: cis OH, 1:1:2:4:6; trans OH, 1 : 1:6 : 6; cis NH?, 1 :2 :6 :6, trans NHz, 1 :8 :6. (13) J. I. Musher, Spectrochim. Acta, 16,835 (1960). (14) N. Muller and W. C. Tosch, J . Chem. Phys., 37,1167 (1962). (15) H. Feltkamp, N. C. Franklin, K. D. Thomas, and W. Briigel, Ann., 683, 64 (1965). (16) H. Feltkamp, N. C. Franklin, W. Kraus, and W. Briigel, ibid., 683, 75 (1965). (17) H. Booth, N. C. Franklin, and G. C. Gidley, Tetrahedron, 21, 1077 (1965). (18) N. S. Bhacca and D. H. Williamson, “Applications of N M R Spectroscopy in Organic Chemistry,” Holden-Day, Inc., San Francisco, Calif., 1964, p 183. (19) H. Booth, Tetrahedron, 22, 615 (1966). (20) E. Eliel and M. H. Gianni, Tetrahedron Letters, 97 (1962).

Journal of the American Chemical Society / 89:l

served an analogous but much less pronounced effect in the 3-isopropylcyclobutanecarboxylateesters. 1 The present results are therefore intelligible only in terms of a relatively homogeneous nonplanar system containing predominantly or exclusively equatorial isopropyl and axial or equatorial OH or NH2. The equilibrium depicted in Figure 1 must lie substantially to the left for both sets of isomers. We believe this to be the first instance of the successful use of a holding group in cyclobutane. A sound basis is thus at hand for further investigations of conformational distinctions in this system. Acknowledgment. The authors are indebted to the Directorate of Chemical Sciences, Air Force Office of Scientific Research, for Grant No. AFOSR-123-63 which provided financial support for this work. We are grateful to Varian Associates for their kindness in running nmr spectra of the amines. (21) I. Lillien and R. A. Doughty, unpublished work. (22) Based in part on the dissertation submitted by R. A. Doughty to the Graduate School of the University of Miami in partial fulfillment of the requirements for the Ph.D. Degree in chemistry.

I. Lillien, R. A. Doughtyzz Deparimeni of Pediatrics, School of Medicine Unicersity of Miami, Miami, Florida Received September 26, I966

Stable Carbonium Ions. XXXIL1 Alkyldicarbonium Ions Sir: Recently we reported the direct observation of alkylenedioxodicarbonium ions (acyl dications). No observation of alkyldicarbonium ions has yet been reported. We now wish to report the direct observation of the first stable alkyldicarbonium ions. When 2,5-dichloro2,5-dimethylhe~ane~and 2,6-dichloro-2,6-dimethylheptane4 were dissolved in S02-SbF6 solution at low temperature, clean, slightly colored solutions resulted which gave the spectra shown in Figures 1 and 2. These spectra are those of the dicarbonium ions I and 11, respectively, in the form of the di-SbFjC1- salts. CH,

CHI

1

I

CHI

CH3

1

I

CH3CCHZCHsFCHa

CH&CH?CHzCH,$CH,

I

I1

B

f2

The dication nature of these ions was deduced from three observations: (1) the nmr spectra, indicative of the ions, in particular the downfield shifts of the methyl and methylene groups adjacent to the positively charged carbon atoms;5 ( 2 ) the long-range coupling of the methyl groups with the adjacent methylene groups through the sp2 and (3) the isolation of dimethyl ethers on quenching solutions of the dications in methanol at - 7 8 O . ’ (1) Part XXXI: G. A. Olah, M. B. Comisarow, and E. Namanworth, J . Am. Chem. SOC.,in press. (2) G. A. Olah and M. B. Comisarow, ibid., 88,3313 (1966). (3) H. A. Bruson, F. W. Grant, and E. Bobko, ibid., 80,3633 (1958). (4) H. Staudinger and W. Widmer, Helv. Chim. Acta, 9, 546 (1926); we substituted glutaryl dichloride for the diester. ( 5 ) G. A. Olah, E. B. Baker, J. C. Evans, W. S. Tolgyesi, J. S. MCIntyre, and I. J. Bastien, J . Am. Chem. Soc., 86, 1360 (1964), and other papers in this series. (6) G. A. Olah and C. U. Pittman, Jr., Advan. Phys. Org. Chem., 4, 303 (1966).

January 4, 1967